Deductive and Empirical Strategies in Personality Testing

Introduction to Deductive and Empirical Strategies

When examining the deductive and empirical strategies used in the construction of structured personality instruments, it is important to clarify what the terms deductive and empirical mean and how they relate to tests designed for psychological purposes. Empirical evidence is that which can be demonstrated or proven and which ultimately exists in the observable world. Deductive reasoning is a form of logic in which individuals establish a basic premise or truth, combine it with other premises for which there is empirical evidence, and then draw conclusions. This type of reasoning determines conclusions based on a top-down approach to logic. These two strategies, which are frequently applied in conjunction with one another, are highly important to the makeup of structured personality tests. Without such strategies, the results of personality instruments would be virtually useless or inconclusive at best.

Defining Deductive Strategies in Personality Test Construction

When examining deductive and empirical strategies for personality instruments, it is pivotal to recognize that these strategies largely pertain to isolating and stratifying different aspects of an individual's personality. Personality tests are constructed for various purposes — some, like the Big Five Factor Personality Model, may be useful in particular situations or yield specific insight, such as identifying which job candidates have a personality aligned with a given organization's culture. Some standardized assessments used for psychological purposes are designed for certain populations (Kaplan, no date, p. 329). They also attempt to categorize and gauge different facets of personality — such as the Myers-Briggs Jungian measures, which categorize personality by factors including extroversion, perceiving information, judging, and orientation to the external world.

In addition to being designed for particular applications and stratifying personality by a series of relevant factors, personality tests are constructed with various types of scales that aid in their deductive and empirical rigor. There are typically primary, secondary, and even tertiary scales, which incorporate a variety of response formats to help accrue data. One of the most useful aspects of this scale variation is that on many tests, such as the 16PF-5, there are scales specifically designed to measure the validity of the test itself (Strack et al., 2000, p. 376). These scales are comprised of a number of different items. It is also beneficial that on some personality tests there are multiple scores, which provide a greater degree of detail and insight into a population's or an individual's results. These tests frequently report means and standard deviations, enabling administrators to identify significant data patterns. Moreover, since these assessments are typically administered individually, they offer the opportunity for an administrator to "observe behavior in a standard situation," which may be "invaluable to an examiner who is trying to understand the unique attributes of a person and interpret the meaning of a test score" (Kaplan, no date, p. 362).

Empirical Strategies for the Construction of Structured Personality Tests

A pivotal strategy for constructing empirically and deductively valid personality tests involves gauging reliability. There are many ways of achieving this objective. One of the most trusted approaches is to utilize samples from various segments of the population in order to gauge the "internal consistency" (Strack et al., 2000, p. 376) of the particular measure being used. By assessing individuals from different segments of a population, test administrators are better able to adjudicate the reliability of the assessment instrument itself. It is also important to administer the test at different points in time to analyze its reliability across occasions. When utilizing specific populations for reliability measurement, it is best to compare the demographics of that group against normative data for the population as a whole. Key factors to consider include gender and racial ethnicity, especially since these variables can introduce bias in other standardized assessments based on aptitude (Suzuki et al., 2000, p. 491). Intelligence is one of the factors measured on numerous personality tests.

In terms of verifying the validity of personality tests, it is crucial to conduct additional studies examining whether their results are empirically sound. There are several aspects of a test's validity that can be reviewed in this way, including the particular factors it identifies — typically organized by scale — as measurable indicators of personality facets, and its predictive value (Strack et al., 2000, p. 378). Finally, one can help assure the validity of a structured personality test by comparing its results with those of other, more established measures. Ideally, there should be meaningful similarity in the results across instruments.

Conclusion

McCrae, R.R., & Costa Jr., P.T. (1989). Rotation to maximize the construct validity of factors in the NEO Personality Inventory. Multivariate Behavioral Research, 24, 107–124.

Strack, K.M., Dunaway, M.H., & Schulenberg, S.E. (2000). Handbook of multicultural assessment. Suzuki, L.A., Ponterotto, J.G. (Ed.). Hoboken: Jossey-Bass.

Suzuki, L.A., Prevost, L., & Short, E.L. (2000). Handbook of multicultural assessment. Suzuki, L.A., Ponterotto, J.G. (Ed.). Hoboken: Jossey-Bass.